An implementation of a garment-fitting simulation system using laser scanned 3D body data Chang-Suk Cho a, *, Jun-Young Park a,1 , Adrian Boeing b , Philip Hingston b,2 a Department of Information Science and Telecom, Hanshin University, 411 Yangsan-dong, Osan-si, Gyeonggi-do, Republic of Korea b School of Computer and Security Science, Edith Cowan University, 2 Bradford St, Mt Lawley, WA 6050, Australia 1. Introduction Currently, online clothing store websites typically only display 2D images of the garments that they offer for sale. They offer no user-interaction or feedback regarding the fit of the garment for a particular user’s body type and size. This paper presents a 3D interactive simulated garment-fitting system that uses 3D laser scanned body data and 2D garment images. Our system introduces a new algorithm for recognizing key characteristic points on the 3D body data, and also a new fitting approach based on using the ‘‘front neck point’’ of a 2D garment image. When implementing a simulated garment-fitting system, there are two main problems to solve: recognition of characteristic points on 3D body data and fitting 2D garment images over the 3D data. To recognize a 3D body surface, Lee and Cho treat the curvature series of each slice of a 3D body as a 2D signal [1]. This approach achieves good results for a standard body, but it can easily be trapped in local minima. An alternative approach uses spin-images of concave and convex parts of 3D body data to recognize characteristic points. However, this method is not robust and not easily generalised to other characteristic points on the body [2]. Therefore, in [3],a template application approach was used to collect features of characteristic points on the 3D surface. It is also important for a simulated garment-fitting system to accurately combine 3D body data and 2D garment images. This requires the geometrical and physical modeling of clothes, collision detection between body data and clothes, and also feedback handling techniques. There are several existing approaches to simulated garment fitting. Seo et al. [4] present a cloth simulator for estimating pressure from tightly fitted clothing, in which physical measurements of pressure were compared to simulated pressure estimates. Magnenat-Thalmann et al. [5] used an avatar model for garment fitting, deforming an avatar to the body size of the consumer. This provides a fitting system that may be useful in an in-house context, but would be time-consuming and inconvenient for the consumer. Cordier et al. [6] presented a well-designed online clothing retail system that uses a web application to manipulate garment- related items, facilitating clothing design, pattern derivation and sizing. This system requires information on the composition of patterns, which is not practical, given the volume of different clothes designs in the market. Furthermore, fitting processes and associated databases are deployed on a centralized system, which is undesirable in terms of security and in terms of computing costs for the retailer. For this reason, we developed a simple system using only 2D images of the front and back of a garment. We propose a Computers in Industry 61 (2010) 550–558 ARTICLE INFO Keywords: 3D body data Moment template Vector angle Garment fitting Simulation Laser scan ABSTRACT We propose a garment-fitting system for an online retail model, which uses a consumer’s 3D body data for garment fitting. This system uses front and back images to model the garment and 3D laser scanned body data to model the body. In order to recognize characteristic points on the 3D body data, a moment template composed of central moments of vector angle sets on 3D body data is proposed, and an implementation combining the 3D data and garment images is reported, which includes modeling, collision detection and feedback handling. The resulting system would support an online retail business model for garments. In the proposed business model, a body data center would scan the body and store the 3D data, the simulated garment-fitting program would be provided to individual consumers for fitting on a home PC, and retail shops would exhibit garment images on their websites. In this proposed model, security of personal data and computational cost will not be an issue. Crown Copyright ß 2010 Published by Elsevier B.V. All rights reserved. * Corresponding author. Tel.: +82 31 379 0652. E-mail addresses: cscho@hs.ac.kr (C.-S. Cho), jypark@eyenix.com (J.-Y. Park), a.boeing@ecu.edu.au (A. Boeing), p.hingston@ecu.edu.au (P. Hingston). 1 Tel.: +82 31 379 0652. 2 Tel.: +61 8 9370 6427. Contents lists available at ScienceDirect Computers in Industry journal homepage: www.elsevier.com/locate/compind 0166-3615/$ – see front matter . Crown Copyright ß 2010 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.compind.2010.03.005